1 데이터
- 성남시
- 전체
- 3구
- 동
- 투표소(?)
1.1 투표율
sungnam_raw <- tibble(연도 = c("2012", "2017", "2022"),
data = list(krvote::election_20121219$투표율,
krvote::election_20170509$투표율,
krvote::election_20220309$투표율) )
sn_casting_raw <- sungnam_raw %>%
mutate(성남 = map(data, ~.x %>% filter(str_detect(구시군명, "성남")))) %>%
select(연도, 성남) %>%
unnest(성남)
sn_casting <- sn_casting_raw %>%
group_by(연도, 구시군명) %>%
summarise(선거인수 = sum(선거인수),
투표수 = sum(투표수)) %>%
ungroup() %>%
mutate(투표율 = 투표수 / 선거인수)
sn_casting# A tibble: 9 x 5
연도 구시군명 선거인수 투표수 투표율
<chr> <chr> <dbl> <dbl> <dbl>
1 2012 성남시분당구 381468 306328 0.803
2 2012 성남시수정구 193613 138079 0.713
3 2012 성남시중원구 206445 147567 0.715
4 2017 성남시분당구 403060 338076 0.839
5 2017 성남시수정구 201105 149414 0.743
6 2017 성남시중원구 198663 146801 0.739
7 2022 성남시분당구 404493 335434 0.829
8 2022 성남시수정구 207998 161006 0.774
9 2022 성남시중원구 186278 142368 0.7641.2 득표수
voting_raw <- tibble(연도 = c("2012", "2017", "2022"),
data = list(krvote::election_20121219$득표율,
krvote::election_20170509$득표율,
krvote::election_20220309$득표율) )
sn_voting <- voting_raw %>%
mutate(data = map(data, ~.x %>% pivot_longer(선거인수:계, names_to = "후보", values_to = "득표수"))) %>%
unnest(data) %>%
filter(str_detect(구시군명, "성남")) %>%
filter(str_detect(후보, "박근혜|문재인|안철수|홍준표|유승민|심상정|이재명|윤석열")) %>%
mutate(정당 = case_when(str_detect(후보, "문재인|이재명") ~ "민주당",
str_detect(후보, "심상정") ~ "정의당",
TRUE ~ "국민의힘")) %>%
group_by(연도, 구시군명, 정당) %>%
summarise(득표수 = sum(득표수)) %>%
ungroup()1.3 득표율
sn_voting_hubo <- voting_raw %>%
mutate(data = map(data, ~.x %>% pivot_longer(선거인수:계, names_to = "후보", values_to = "득표수"))) %>%
unnest(data) %>%
filter(str_detect(구시군명, "성남")) %>%
filter(str_detect(후보, "박근혜|문재인|안철수|홍준표|유승민|심상정|이재명|윤석열")) %>%
mutate(정당 = case_when(str_detect(후보, "문재인|이재명") ~ "민주당",
str_detect(후보, "심상정") ~ "정의당",
TRUE ~ "국민의힘")) %>%
group_by(연도, 구시군명, 정당) %>%
summarise(득표수 = sum(득표수)) %>%
ungroup()
sn_voting_total <- voting_raw %>%
mutate(data = map(data, ~.x %>% pivot_longer(선거인수:계, names_to = "후보", values_to = "득표수"))) %>%
unnest(data) %>%
filter(str_detect(구시군명, "성남")) %>%
filter(str_detect(후보, "투표수")) %>%
group_by(연도, 구시군명) %>%
summarise(투표수 = sum(득표수)) %>%
ungroup()
sn_voting_pcnt <- sn_voting_hubo %>%
left_join(sn_voting_total) %>%
mutate(득표율 = 득표수 / 투표수)
sn_voting_pcnt# A tibble: 24 x 6
연도 구시군명 정당 득표수 투표수 득표율
<chr> <chr> <chr> <dbl> <dbl> <dbl>
1 2012 성남시분당구 국민의힘 162103 306328 0.529
2 2012 성남시분당구 민주당 142315 306328 0.465
3 2012 성남시수정구 국민의힘 62526 138079 0.453
4 2012 성남시수정구 민주당 74640 138079 0.541
5 2012 성남시중원구 국민의힘 68309 147567 0.463
6 2012 성남시중원구 민주당 78261 147567 0.530
7 2017 성남시분당구 국민의힘 175723 338076 0.520
8 2017 성남시분당구 민주당 139853 338076 0.414
9 2017 성남시분당구 정의당 20415 338076 0.0604
10 2017 성남시수정구 국민의힘 74462 149414 0.498
# ... with 14 more rows2 시각화
2.1 투표율
library(ggrepel)
sn_casting_total <- sn_casting_raw %>%
group_by(연도) %>%
summarise(선거인수 = sum(선거인수),
투표수 = sum(투표수)) %>%
ungroup() %>%
mutate(투표율 = 투표수 / 선거인수) %>%
mutate(구시군명 = "전체")
sn_casting_g <- sn_casting %>%
bind_rows(sn_casting_total) %>%
mutate(구시군명 = str_remove(구시군명, "성남시")) %>%
mutate(구시군명 = factor(구시군명, levels = c("전체", "분당구", "수정구", "중원구"))) %>%
ggplot(aes(x = 연도, y = 투표율, group = 구시군명)) +
geom_line(aes(color = 구시군명)) +
geom_point(aes(color = 구시군명), size = 5) +
scale_y_continuous(labels = scales::percent) +
labs(x="",
title = "성남시 대통령선거 투표율") +
theme_bw(base_family = "NanumBarunPen") +
theme(
text = element_text(family = "NanumGothic"),
plot.tag.position = c(0.85, .97),
legend.position = "top",
legend.title=element_text(size=15),
legend.text=element_text(size=13),
axis.title.y = element_text(size = rel(1.5), colour = "black", family = "NanumBarunpen", face="bold"),
strip.text.x = element_text(size = rel(2.5), colour = "black", family = "NanumMyeongjo", face="bold"),
axis.text.y = element_text(size = rel(1.5), colour = "gray35", family = "NanumBarunpen", face="bold"),
axis.text.x = element_text(size = rel(1.4), colour = "black", family = "NanumBarunpen", face="bold"),
strip.background=element_rect(fill="gray95"),
plot.title=element_text(size=25, face="bold", family = "NanumBarunpen"),
plot.subtitle=element_text(face="bold", size=17, colour="grey10", family = "NanumBarunpen")) +
geom_text_repel(aes(label = scales::percent(투표율, accuracy =0.1)), hjust = 1, size = 5 ) +
scale_color_manual(values = c("전체" = "black",
"분당구" = "red",
"수정구" = "blue",
"중원구" = "green"))
sn_casting_g
ragg::agg_png("fig/sn_casting_g.png", width = 297, height = 210, units = "mm", res = 300, scaling = 0.7)
sn_casting_g
dev.off()
2.2 득표수
sn_voting_count_overall_g <- sn_voting %>%
group_by(연도, 정당) %>%
summarise(득표수 = sum(득표수)) %>%
ggplot(aes(x = 연도, y = 득표수, color = 정당, group = 정당)) +
geom_point() +
geom_line() +
geom_point(size = 5) +
scale_y_continuous(labels = scales::comma) +
labs(x="",
title = "성남시 대통령선거 득표수",
caption = "2017년 대선 국민의힘 득표수 = 홍준표 + 안철수 + 유승민") +
theme_bw(base_family = "NanumBarunPen") +
theme(
text = element_text(family = "NanumGothic"),
plot.tag.position = c(0.85, .97),
legend.position = "top",
legend.title=element_text(size=15),
legend.text=element_text(size=13),
axis.title.y = element_text(size = rel(1.5), colour = "black", family = "NanumBarunpen", face="bold"),
strip.text.x = element_text(size = rel(2.5), colour = "black", family = "NanumBarunpen", face="bold"),
axis.text.y = element_text(size = rel(1.5), colour = "gray35", family = "NanumBarunpen", face="bold"),
axis.text.x = element_text(size = rel(1.4), colour = "black", family = "NanumBarunpen", face="bold"),
strip.background=element_rect(fill="gray95"),
plot.title=element_text(size=25, face="bold", family = "NanumBarunpen"),
plot.subtitle=element_text(face="bold", size=17, colour="grey10", family = "NanumBarunpen")) +
geom_text_repel(aes(label = scales::comma(득표수)), size = 5 ) +
scale_color_manual(values = c("국민의힘" = "red",
"민주당" = "blue",
"정의당" = "green"))
sn_voting_count_overall_g
ragg::agg_png("fig/sn_voting_count_overall_g", width = 297, height = 210, units = "mm", res = 300, scaling = 0.7)
sn_voting_count_overall_g
dev.off()2.2.1 3구
sn_voting_count_g <- sn_voting %>%
ggplot(aes(x = 연도, y = 득표수, color = 정당, group = 정당)) +
geom_point() +
geom_line() +
facet_wrap( ~ 구시군명) +
geom_point(size = 5) +
scale_y_continuous(labels = scales::comma) +
labs(x="",
title = "성남시 대통령선거 득표수",
caption = "2017년 대선 국민의힘 득표수 = 홍준표 + 안철수 + 유승민") +
theme_bw(base_family = "NanumBarunPen") +
theme(
text = element_text(family = "NanumGothic"),
plot.tag.position = c(0.85, .97),
legend.position = "top",
legend.title=element_text(size=15),
legend.text=element_text(size=13),
axis.title.y = element_text(size = rel(1.5), colour = "black", family = "NanumBarunpen", face="bold"),
strip.text.x = element_text(size = rel(2.5), colour = "black", family = "NanumBarunpen", face="bold"),
axis.text.y = element_text(size = rel(1.5), colour = "gray35", family = "NanumBarunpen", face="bold"),
axis.text.x = element_text(size = rel(1.4), colour = "black", family = "NanumBarunpen", face="bold"),
strip.background=element_rect(fill="gray95"),
plot.title=element_text(size=25, face="bold", family = "NanumBarunpen"),
plot.subtitle=element_text(face="bold", size=17, colour="grey10", family = "NanumBarunpen")) +
geom_text(aes(label = scales::comma(득표수)), vjust = -1.0, size = 5 ) +
scale_color_manual(values = c("국민의힘" = "red",
"민주당" = "blue",
"정의당" = "green"))
sn_voting_count_g
ragg::agg_png("fig/sn_voting_count_g.png", width = 297, height = 210, units = "mm", res = 300, scaling = 0.7)
sn_voting_count_g
dev.off()
2.3 득표율
2.3.1 성남시 전체
sn_voting_pcnt_overall_g <- sn_voting_pcnt %>%
group_by(연도, 정당) %>%
summarise(득표수 = sum(득표수),
투표수 = sum(투표수)) %>%
ungroup() %>%
mutate(득표율 = 득표수 / 투표수) %>%
ggplot(aes(x = 연도, y = 득표율, color = 정당, group = 정당)) +
geom_point() +
geom_line() +
geom_point(size = 5) +
scale_y_continuous(labels = scales::percent) +
labs(x="",
title = "성남시 대통령선거 득표율",
caption = "2017년 대선 국민의힘 득표율(%) = 홍준표 + 안철수 + 유승민") +
theme_bw(base_family = "NanumBarunPen") +
theme(
text = element_text(family = "NanumGothic"),
plot.tag.position = c(0.85, .97),
legend.position = "top",
legend.title=element_text(size=15),
legend.text=element_text(size=13),
axis.title.y = element_text(size = rel(1.5), colour = "black", family = "NanumBarunpen", face="bold"),
strip.text.x = element_text(size = rel(2.5), colour = "black", family = "NanumBarunpen", face="bold"),
axis.text.y = element_text(size = rel(1.5), colour = "gray35", family = "NanumBarunpen", face="bold"),
axis.text.x = element_text(size = rel(1.4), colour = "black", family = "NanumBarunpen", face="bold"),
strip.background=element_rect(fill="gray95"),
plot.title=element_text(size=25, face="bold", family = "NanumBarunpen"),
plot.subtitle=element_text(face="bold", size=17, colour="grey10", family = "NanumBarunpen")) +
geom_text_repel(aes(label = scales::percent(득표율, accuracy = 0.1)), size = 5 ) +
scale_color_manual(values = c("국민의힘" = "red",
"민주당" = "blue",
"정의당" = "green"))
sn_voting_pcnt_overall_g
ragg::agg_png("fig/sn_voting_pcnt_overall_g.png", width = 297, height = 210, units = "mm", res = 300, scaling = 0.7)
sn_voting_pcnt_overall_g
dev.off()
2.3.2 3구
sn_voting_pcnt_g <- sn_voting_pcnt %>%
ggplot(aes(x = 연도, y = 득표율, color = 정당, group = 정당)) +
geom_point() +
geom_line() +
facet_wrap( ~ 구시군명) +
geom_point(size = 5) +
scale_y_continuous(labels = scales::percent) +
labs(x="",
title = "성남시 대통령선거 득표율",
caption = "2017년 대선 국민의힘 득표율(%) = 홍준표 + 안철수 + 유승민") +
theme_bw(base_family = "NanumBarunPen") +
theme(
text = element_text(family = "NanumGothic"),
plot.tag.position = c(0.85, .97),
legend.position = "top",
legend.title=element_text(size=15),
legend.text=element_text(size=13),
axis.title.y = element_text(size = rel(1.5), colour = "black", family = "NanumBarunpen", face="bold"),
strip.text.x = element_text(size = rel(2.5), colour = "black", family = "NanumBarunpen", face="bold"),
axis.text.y = element_text(size = rel(1.5), colour = "gray35", family = "NanumBarunpen", face="bold"),
axis.text.x = element_text(size = rel(1.4), colour = "black", family = "NanumBarunpen", face="bold"),
strip.background=element_rect(fill="gray95"),
plot.title=element_text(size=25, face="bold", family = "NanumBarunpen"),
plot.subtitle=element_text(face="bold", size=17, colour="grey10", family = "NanumBarunpen")) +
geom_text(aes(label = scales::percent(득표율, accuracy = 0.1)), vjust = -1.0, size = 5 ) +
scale_color_manual(values = c("국민의힘" = "red",
"민주당" = "blue",
"정의당" = "green"))
sn_voting_pcnt_g
ragg::agg_png("fig/sn_voting_pcnt_g.png", width = 297, height = 210, units = "mm", res = 300, scaling = 0.7)
sn_voting_pcnt_g
dev.off()
데이터 과학자 이광춘 저작
kwangchun.lee.7@gmail.com